MAP: Unleashing Hybrid Mamba-Transformer Vision Backbone's Potential with Masked Autoregressive Pretraining

• URL: http://arxiv.org/abs/2410.00871v1
• Date: Tue, 1 Oct 2024 17:05:08 GMT
• Title: MAP: Unleashing Hybrid Mamba-Transformer Vision Backbone's Potential with Masked Autoregressive Pretraining
• Authors: Yunze Liu, Li Yi,
• Abstract summary: We propose Masked Autoregressive Pretraining (MAP) to pretrain a hybrid Mamba-Transformer vision backbone network. We show that both the pure Mamba architecture and the hybrid Mamba-Transformer vision backbone network pretrained with MAP significantly outperform other pretraining strategies.
• Score: 23.37555991996508
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Mamba has achieved significant advantages in long-context modeling and autoregressive tasks, but its scalability with large parameters remains a major limitation in vision applications. pretraining is a widely used strategy to enhance backbone model performance. Although the success of Masked Autoencoder in Transformer pretraining is well recognized, it does not significantly improve Mamba's visual learning performance. We found that using the correct autoregressive pretraining can significantly boost the performance of the Mamba architecture. Based on this analysis, we propose Masked Autoregressive Pretraining (MAP) to pretrain a hybrid Mamba-Transformer vision backbone network. This strategy combines the strengths of both MAE and Autoregressive pretraining, improving the performance of Mamba and Transformer modules within a unified paradigm. Additionally, in terms of integrating Mamba and Transformer modules, we empirically found that inserting Transformer layers at regular intervals within Mamba layers can significantly enhance downstream task performance. Experimental results show that both the pure Mamba architecture and the hybrid Mamba-Transformer vision backbone network pretrained with MAP significantly outperform other pretraining strategies, achieving state-of-the-art performance. We validate the effectiveness of the method on both 2D and 3D datasets and provide detailed ablation studies to support the design choices for each component.

Related papers

• TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba [88.31117598044725]
  We explore cross-architecture training to transfer the ready knowledge in existing Transformer models to alternative architecture Mamba, termed TransMamba. Our approach employs a two-stage strategy to expedite training new Mamba models, ensuring effectiveness in across uni-modal and cross-modal tasks. For cross-modal learning, we propose a cross-Mamba module that integrates language awareness into Mamba's visual features, enhancing the cross-modal interaction capabilities of Mamba architecture.
  arXiv  Detail & Related papers  (2025-02-21T01:22:01Z)
• MatIR: A Hybrid Mamba-Transformer Image Restoration Model [95.17418386046054]
  We propose a Mamba-Transformer hybrid image restoration model called MatIR. MatIR cross-cycles the blocks of the Transformer layer and the Mamba layer to extract features. In the Mamba module, we introduce the Image Inpainting State Space (IRSS) module, which traverses along four scan paths.
  arXiv  Detail & Related papers  (2025-01-30T14:55:40Z)
• MobileMamba: Lightweight Multi-Receptive Visual Mamba Network [51.33486891724516]
  Previous research on lightweight models has primarily focused on CNNs and Transformer-based designs. We propose the MobileMamba framework, which balances efficiency and performance. MobileMamba achieves up to 83.6% on Top-1, surpassing existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-11-24T18:01:05Z)
• MambaPEFT: Exploring Parameter-Efficient Fine-Tuning for Mamba [0.5530212768657544]
  Mamba, a State Space Model (SSM)-based model, has attracted attention as a potential alternative to Transformers. We investigate the effectiveness of existing PEFT methods for Transformers when applied to Mamba. We propose new Mamba-specific PEFT methods that leverage the distinctive structure of Mamba.
  arXiv  Detail & Related papers  (2024-11-06T11:57:55Z)
• MaskMamba: A Hybrid Mamba-Transformer Model for Masked Image Generation [63.73137438677585]
  MaskMamba is a novel hybrid model that combines Mamba and Transformer architectures. It achieves a remarkable $54.44%$ improvement in inference speed at a resolution of $2048times 2048$ over Transformer.
  arXiv  Detail & Related papers  (2024-09-30T04:28:55Z)
• Famba-V: Fast Vision Mamba with Cross-Layer Token Fusion [10.854742185190482]
  Mamba and Vision Mamba (Vim) models have shown their potential as an alternative to methods based on Transformer architecture. This work introduces Fast Mamba for Vision (Famba-V), a cross-layer token fusion technique to enhance the training efficiency of Vim models.
  arXiv  Detail & Related papers  (2024-09-15T18:02:26Z)
• ReMamba: Equip Mamba with Effective Long-Sequence Modeling [50.530839868893786]
  We propose ReMamba, which enhances Mamba's ability to comprehend long contexts. ReMamba incorporates selective compression and adaptation techniques within a two-stage re-forward process.
  arXiv  Detail & Related papers  (2024-08-28T02:47:27Z)
• The Mamba in the Llama: Distilling and Accelerating Hybrid Models [76.64055251296548]
  We show that it is feasible to distill large Transformers into linear RNNs by reusing the linear projection weights from attention layers with academic GPU resources. The resulting hybrid model, which incorporates a quarter of the attention layers, achieves performance comparable to the original Transformer in chat benchmarks.
  arXiv  Detail & Related papers  (2024-08-27T17:56:11Z)
• Transformers to SSMs: Distilling Quadratic Knowledge to Subquadratic Models [92.36510016591782]
  We present a method that is able to distill a pretrained Transformer architecture into alternative architectures such as state space models (SSMs) Our method, called MOHAWK, is able to distill a Mamba-2 variant based on the Phi-1.5 architecture using only 3B tokens and a hybrid version (Hybrid Phi-Mamba) using 5B tokens. Despite using less than 1% of the training data typically used to train models from scratch, Phi-Mamba boasts substantially stronger performance compared to all past open-source non-Transformer models.
  arXiv  Detail & Related papers  (2024-08-19T17:48:11Z)
• MambaMIM: Pre-training Mamba with State Space Token-interpolation [14.343466340528687]
  We introduce a generative self-supervised learning method for Mamba (MambaMIM) based on Selective Structure State Space Sequence Token-interpolation (S6T) MambaMIM can be used on any single or hybrid Mamba architectures to enhance the Mamba long-range representation capability.
  arXiv  Detail & Related papers  (2024-08-15T10:35:26Z)
• An Empirical Study of Mamba-based Pedestrian Attribute Recognition [15.752464463535178]
  This paper designs and adapts Mamba into two typical PAR frameworks, text-image fusion approach and pure vision Mamba multi-label recognition framework. It is found that interacting with attribute tags as additional input does not always lead to an improvement, specifically, Vim can be enhanced, but VMamba cannot. These experimental results indicate that simply enhancing Mamba with a Transformer does not always lead to performance improvements but yields better results under certain settings.
  arXiv  Detail & Related papers  (2024-07-15T00:48:06Z)
• MambaVision: A Hybrid Mamba-Transformer Vision Backbone [54.965143338206644]
  We propose a novel hybrid Mamba-Transformer backbone, denoted as MambaVision, which is specifically tailored for vision applications. Our core contribution includes redesigning the Mamba formulation to enhance its capability for efficient modeling of visual features. We conduct a comprehensive ablation study on the feasibility of integrating Vision Transformers (ViT) with Mamba.
  arXiv  Detail & Related papers  (2024-07-10T23:02:45Z)
• Autoregressive Pretraining with Mamba in Vision [45.25546594814871]
  This paper shows that Mamba's visual capability can be significantly enhanced through autoregressive pretraining. Performance-wise, autoregressive pretraining equips the Mamba architecture with markedly higher accuracy. Our huge-size Mamba attains 85.0% ImageNet accuracy when finetuned with $384times384$ inputs.
  arXiv  Detail & Related papers  (2024-06-11T17:58:34Z)
• Demystify Mamba in Vision: A Linear Attention Perspective [72.93213667713493]
  Mamba is an effective state space model with linear computation complexity. We show that Mamba shares surprising similarities with linear attention Transformer. We propose a Mamba-Like Linear Attention (MLLA) model by incorporating the merits of these two key designs into linear attention.
  arXiv  Detail & Related papers  (2024-05-26T15:31:09Z)
• Is Mamba Capable of In-Context Learning? [63.682741783013306]
  State of the art foundation models such as GPT-4 perform surprisingly well at in-context learning (ICL) This work provides empirical evidence that Mamba, a newly proposed state space model, has similar ICL capabilities.
  arXiv  Detail & Related papers  (2024-02-05T16:39:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.